Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The sound of the piano was intrusive. It was going on and on. Mr. Peterson, the new neighbor, was playing again. Yet, I found myself captivated, not annoyed. A strange warmth spread through me. I imagined him there, hunched over the keys, lost in the music.

He had that boyish charm and I’d heard him greeting the other neighbors in the hall. It was a stark contrast to my usual quiet, introverted life. I'd even started baking, just in case he'd stop by.

Instead of writing a note, as I usually would, I found myself humming along, a silly little smile on my face. The music was like a beacon, drawing me in.

Emotion: infatuated

Cluster: Compassion / Love
PC1 (Valence): 1.61 Positive
PC2 (Disposition): 0.53

Role in Research

This story is one of 1,000 stories generated for the emotion infatuated. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 infatuated stories, after denoising with neutral dialogue baselines, produces the infatuated emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the infatuated vector is projected through the unembedding matrix.

Promoted:
de0.525
šŸ˜0.421
🤩0.418
šŸ˜0.410
ą¤‰ą¤øą„‡0.397
Suppressed:
que-0.427
怨-0.330
a-0.311
ė„-0.307
ꄤ-0.299